Mechanical turntable damage monitoring method and system based on acoustic emission technology combined with svm damage classification

Abstract

The application discloses a mechanical rotary table damage monitoring method and system based on acoustic emission technology combined with SVM damage classification, belongs to the technical field of mechanical indexing rotary table damage detection, and comprises the following steps: installing an acoustic emission sensor on a mechanical rotary table to be detected; acquiring acoustic emission waveforms collected by the acoustic emission sensor in real time, extracting AE events based on original waveforms; calculating acoustic emission characteristic parameters of each AE event and constructing a data set; collecting AE signals of different damages in different regions of the mechanical rotary table, and taking the acoustic emission characteristic parameters of the AE events as a training set; taking characteristic parameters of AE signals of different damages in the same region detected subsequently as a test set; inputting the training set and the test set into SVM to obtain a result; and monitoring the rotary table damage condition by using the accuracy of the SVM according to the result of the SVM. The application can dynamically monitor the damage of a mechanical indexing rotary table in operation and can prevent faults in advance.

Description

Technical Field

[0001] This invention relates to the field of mechanical indexing turntable damage detection technology, specifically to a mechanical turntable damage monitoring method and system based on acoustic emission technology combined with SVM damage classification. Background Technology

[0002] Mechanical rotary tables operate continuously on industrial production lines for extended periods, and inadequate routine maintenance can easily lead to wear and even breakage of the spindle. Therefore, damage monitoring of mechanical rotary table spindles is of paramount importance to prevent these issues from affecting product quality or even causing serious safety accidents due to severe damage.

[0003] Traditional methods for inspecting rotating machinery mainly include vibration testing and temperature testing. Vibration testing analyzes and processes vibration signals to extract normal and fault characteristics. However, vibration testing requires mounting sensors on the inspected component, is suitable for detecting known fault types, and is not sensitive enough to early material damage. Temperature-based testing monitors temperature changes in the equipment; sudden abnormal temperature changes during operation indicate a potential fault. This method is only applicable to certain fault types and has low sensitivity to some mechanical problems. Both methods are insensitive to equipment faults and cannot effectively address the problem of online inspection. Therefore, a non-destructive testing method capable of dynamically inspecting mechanical indexing rotary tables is needed.

[0004] Acoustic emission (AE), also known as stress wave emission, is the phenomenon where energy is released in the form of elastic waves when a material deforms or fractures under external or internal forces. AE technology can not only dynamically monitor stress changes inside and on the surface of materials, but also perform early damage identification on various materials such as metals, rocks, and composite materials. In recent years, acoustic emission technology has been increasingly widely used in damage detection of industrial equipment such as bearings, speed reducers, and rotating machinery. Summary of the Invention

[0005] The present invention addresses the problem that existing technical solutions are too simplistic and provides a solution that is significantly different from existing technologies. It mainly provides a mechanical turntable damage monitoring method and system based on acoustic emission technology combined with SVM damage classification, in order to solve the technical problem that traditional detection methods mentioned in the background cannot achieve online dynamic monitoring.

[0006] The technical solution adopted by the present invention to solve the above-mentioned technical problems is as follows:

[0007] A mechanical turntable damage monitoring method based on acoustic emission technology combined with SVM damage classification includes the following steps:

[0008] S1. Install an acoustic emission sensor on the mechanical turntable to be tested;

[0009] S2. Acquire the acoustic emission waveform collected by the acoustic emission sensor in real time, and extract the AE event based on the original waveform;

[0010] S3. Calculate the acoustic emission characteristic parameters for each AE event and construct the dataset;

[0011] S4. Collect AE signals of different damages in different areas of the mechanical turntable, and calculate the acoustic emission characteristic parameters of each AE event to form a training set for different damages.

[0012] S5. Calculate the acoustic emission characteristic parameters corresponding to the AE signals of different damages detected in the same area, and use them as the test set.

[0013] S6. Input the above training set and test set into the SVM to obtain the results;

[0014] S7. Based on the results of SVM, use the accuracy of SVM to monitor the damage to the turntable.

[0015] Furthermore, in step S2, the AE event is identified based on a preset threshold voltage and duration.

[0016] Further, in step S2, firstly, data points in the original waveform that exceed the threshold voltage are filtered out, and the positions of data points with an interval greater than the preset duration between two adjacent data points are determined; then, the start and end points of each AE event are recorded, and the duration of the AE event is calculated.

[0017] Furthermore, in step S3, the acoustic emission characteristic parameters include amplitude, ring count, margin, kurtosis, and centroid frequency.

[0018] Further, in step S3, the formula for calculating the acoustic emission characteristic parameters is:

[0019]

[0020]

[0021]

[0022]

[0023]

[0024]

[0025] Among them, X i This represents a time-domain voltage signal, where AM represents amplitude and X represents amplitude. RMgn represents the root magnitude, and X represents the margin index. max Indicates the maximum value of the signal. Let represent the standard deviation, u represent the signal mean, Kurt represent the kurtosis index, FC represent the centroid frequency, f represent the frequency, and P(f) represent the power spectral density of the signal.

[0026] Furthermore, in step S3, for the i-th AE event, the dataset is... Where AM represents amplitude, RC represents ring count, Mgn represents margin, Kurt represents kurtosis, and FC represents centroid frequency. This indicates the total number of AE events extracted.

[0027] Furthermore, in step S7, the number of samples that should have been in a normal condition but were predicted as minor damage and the number of samples that should have been minor damage but were predicted as severe damage are counted. The accuracy is calculated by combining these numbers with the total number of samples in the test set.

[0028] This invention also provides a mechanical turntable damage monitoring system based on acoustic emission technology combined with SVM damage classification, used to implement the steps of the above-mentioned mechanical turntable damage monitoring method based on acoustic emission technology combined with SVM damage classification. The damage monitoring system includes an acoustic emission signal acquisition module, an acoustic emission signal processing module, and a damage monitoring module; wherein:

[0029] The acoustic emission signal acquisition module includes an acoustic emission sensor, which is used to acquire the AE signal released during the operation of the mechanical turntable under test, and convert the sound signal into an electrical signal;

[0030] The acoustic emission signal processing module includes a preamplifier and an acoustic emitter. The output of the acoustic emission sensor is connected to the input of the preamplifier via a signal cable. The acoustic emitter is used to filter the signal, perform analog-to-digital conversion, and generate a waveform file for transmission to the host.

[0031] The damage monitoring module includes a host computer, which is used to acquire the original acoustic emission waveform file and perform monitoring according to the steps of the mechanical turntable damage monitoring method based on acoustic emission technology combined with SVM damage classification.

[0032] Furthermore, the damage monitoring module is also equipped with an indicator light alarm device, which is connected to the host, and displays the different degrees of damage to the mechanical turntable under test through different states of the indicator lights.

[0033] Furthermore, the acoustic emission sensor is a piezoelectric ceramic sensor with a bandwidth frequency range of 50kHz-400kHz and a resonant frequency of 150kHz; the preamplifier has an amplification factor of 40dB.

[0034] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0035] (1) This invention is based on acoustic emission technology. It collects acoustic emission (AE) signals of different damages in different areas of a mechanical turntable, uses the corresponding AE feature parameters as a training set, and uses the AE signals of different damages detected at various locations as a test set. The training set and the test set are then input into an SVM (Structured Vacuum Instrument). By comparing the predicted and actual values ​​through the SVM classification results, the accuracy of the SVM classification can be obtained. When the accuracy rate decreases significantly, it indicates that the mechanical turntable is severely damaged and requires timely maintenance. Compared with traditional methods, this invention can perform online dynamic damage monitoring of a running mechanical indexing turntable, promptly monitor the time and extent of equipment damage, and thus prevent failures in advance.

[0036] (2) The system provided by this invention is simple to deploy and has a wide range of applications. The acoustic emission sensor is easy to install, has high detection sensitivity, and can be installed at different detection locations according to requirements; it is suitable for various occasions: new machine detection, operation detection, shutdown detection, and can detect early damage with high accuracy.

[0037] The present invention will be explained in detail below with reference to the accompanying drawings and specific embodiments. Attached Figure Description

[0038] Figure 1 A flowchart of the damage monitoring method provided by the present invention;

[0039] Figure 2 This is a structural block diagram of the mechanical turntable damage monitoring system in this invention;

[0040] Figure 3 shows the time domain diagrams of a single complete rotation cycle and different damages in the embodiment.

[0041] Figure 4 The image shows the result obtained by inputting the feature parameters of the second acoustic emission experiment in the embodiment into the SVM.

[0042] Figure 5 The image shows the result obtained by inputting the feature parameters of the third acoustic emission experiment in the embodiment into the SVM. Detailed Implementation

[0043] To facilitate understanding of the present invention, a more comprehensive description of the present invention will be given below with reference to the accompanying drawings, which illustrate several embodiments of the present invention. However, the present invention can be implemented in different forms and is not limited to the embodiments described in the text. Rather, these embodiments are provided to make the disclosure of the present invention more thorough and complete.

[0044] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly associated with those skilled in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0045] Please refer to the attached document carefully. Figure 2 A mechanical turntable damage monitoring system based on acoustic emission technology combined with SVM damage classification includes: an acoustic emission signal acquisition module, an acoustic emission signal processing module, and a damage monitoring module. Wherein:

[0046] The acoustic emission signal acquisition module is completed by an acoustic emission sensor. The acoustic emission sensor is installed on the mechanical turntable to be tested to collect the AE signal released during its operation and convert the sound signal into an electrical signal.

[0047] The acoustic emission signal processing module consists of a preamplifier and an acoustic emission unit. The output of the acoustic emission sensor is connected to the input of the preamplifier via a signal cable. Because the output of the acoustic emission signal, after being converted into an electrical signal, is often quite small, this weak signal may be difficult to distinguish from noise signals after long-distance transmission. Therefore, a preamplifier can improve the signal-to-noise ratio. The acoustic emission unit can filter the signal, perform analog-to-digital conversion, and generate waveform files for transmission to the host computer.

[0048] The damage monitoring module transmits the acquired raw acoustic emission waveform files to the data analysis software MATLAB on the host computer, and then performs monitoring according to the steps of the mechanical turntable damage monitoring method based on acoustic emission technology combined with SVM damage classification.

[0049] The system also includes a traffic light alarm device connected to the main unit.

[0050] (1) Calculate the accuracy of damage classification. The accuracy of SVM damage classification does not change much, that is, the rate of normal damage becoming minor damage is ≤10%, and the rate of normal and minor damage becoming serious damage is ≤5%. The signal light does not light up.

[0051] (2) When the accuracy rate decreases significantly, and the original normal signal becomes a slightly damaged signal >10%, and becomes a seriously damaged signal ≤5%, the indicator light will flash.

[0052] (3) When the accuracy rate decreases significantly, and the originally normal signal becomes slightly damaged (>10%) and becomes severely damaged (>5%), the indicator light will remain on.

[0053] A method for monitoring damage to a mechanical turntable based on acoustic emission technology combined with SVM damage classification using the above system includes the following steps:

[0054] S1. Install at least one acoustic emission sensor on the mechanical turntable to be tested;

[0055] S2. Acquire acoustic emission waveforms from the acoustic emission sensor in real time, and extract AE events based on the original waveforms. Set the threshold voltage (h) and duration (T) in the data analysis software; filter data points in the original waveform that exceed the threshold voltage, and determine the positions of data points where the interval between two adjacent impacts is greater than the duration T; if all intervals of the found data points are less than the set duration, they are considered as one AE event. Then record the start and end points of each AE event and count the duration of the AE events. For AE events with too short a duration, they can be reduced based on the preset duration (analyze the extracted AE events, and if AE events with undesirable or non-standard waveform distributions are found, set the preset duration again for secondary filtering).

[0056] Among them, threshold voltage refers to a preset voltage value, and only when the amplitude of the AE signal exceeds this voltage value can it be detected; duration refers to the duration of the AE event; impact refers to any signal that exceeds the threshold and causes the system channel to collect data; preset duration refers to the duration of the ideal AE event for secondary screening.

[0057] S3. Calculate and statistically analyze the acoustic emission characteristic parameters of each AE event, including five indicators: amplitude (AM), ring count (RC), margin (Mgn), kurtosis (Kurt), and center of gravity frequency (FC).

[0058] Let x1, x2, ..., xt be a set of discrete data obtained by sampling a certain signal x(t). i , ...x N N is the signal length. The standard deviation is given by the formula:

[0059]

[0060]

[0061]

[0062]

[0063]

[0064]

[0065] Among them, X i This represents a time-domain voltage signal, where AM represents amplitude and X represents amplitude. R Mgn represents the root magnitude, and X represents the margin index. maxIndicates the maximum value of the signal. Let represent the standard deviation, u represent the signal mean, Kurt represent the kurtosis index, FC represent the centroid frequency, f represent the frequency, and P(f) represent the power spectral density of the signal.

[0066] For the i-th AE event, construct a dataset consisting of these 5 metrics. (m represents the total number of AE events extracted);

[0067] S4. Multiple experiments were conducted to collect AE signals of different damages in different areas of the mechanical turntable, and the above 5 characteristic parameters were calculated to form a training set for different damages.

[0068] S5. Calculate the acoustic emission characteristic parameters corresponding to the AE signals of different damages detected in the same area, and use them as the test set.

[0069] S6. Input the above training set and test set into SVM (Support Vector Machine) to obtain the results;

[0070] S7. Based on the results of SVM, use the accuracy of SVM to monitor the damage of the turntable under normal conditions and minor damage.

[0071] The accuracy is calculated as follows: count the number of samples A that should have been normal but were predicted as minor injury and the number of samples B that should have been minor injury but were predicted as serious injury. Combine this with the total number of test set samples C. The overall accuracy is calculated as: (CBA) / C*100%.

[0072] If a monitoring session reveals a large number of instances where what should have been normal conditions turned into minor damage, or where what should have been minor damage turned into severe damage, resulting in a significant decrease in classification accuracy, then the mechanical turntable is considered to require timely maintenance.

[0073] Example: The technical method proposed in this invention is applied to a mechanical turntable, model EDX1370, on the rear floor panel line of an automobile company. This turntable consists of a turntable plane, bearings, a housing, a main shaft, a main shaft cam, a main shaft guide rail, and a motor. It has three working states in one work cycle: 180° counterclockwise rotation, standby, and 180° clockwise rotation. The mechanical turntable completes the transfer operation of automotive welding parts through clockwise and counterclockwise rotation.

[0074] The acoustic emission instrument used is the DS5-16C acoustic emission instrument manufactured by Beijing Ruandao. This instrument uses a USB 3.0 interface, has a sampling rate of up to 10 MHz, and features 16 acquisition channels. Each channel's signal sampling frequency is set to 2.5 MHz, and the preamplifier gain is set to 40 dB. The acoustic emission sensor is an RS-2A piezoelectric ceramic sensor with a bandwidth frequency range of 50 kHz to 400 kHz and a resonant frequency of 150 kHz. The acoustic emission sensor is mounted on the spindle of the mechanical turntable.

[0075] 1. Acoustic emission experiments were conducted on the above-mentioned mechanical turntable. The results of a single complete rotation cycle and time-domain plots corresponding to different damage types are shown in Figure 3.

[0076] As shown in Figure 3(a), many sudden signals of varying degrees appeared in the entire AE signal. Through comparative research on the mechanics and the signal, the image captured from the minor damage stage is shown in Figure 3(c). Compared to the normal AE signal in Figure 3(b), the image of minor damage exhibits a continuous AE signal with a relatively high amplitude. In contrast, the AE signal of severe turntable damage in Figure 3(d) shows a sudden AE signal with particularly significant amplitude changes.

[0077] 2. Conduct acoustic emission experiments on the above-mentioned mechanical turntable at three different time periods (each two adjacent periods are 3 months apart):

[0078] The AE signals during the process of the mechanical indexing turntable rotating 180° counterclockwise and 180° clockwise were collected. In the MATLAB analysis software, the threshold voltage h=200mV and the duration T=500uS were set to extract the AE events and calculate five indicators including amplitude (AM), ring count (RC), margin (Mgn), kurtosis (Kurt) and center of gravity frequency (FC).

[0079] The feature parameters corresponding to different damage locations in the first experiment were used as the training set; the feature parameters calculated for the same damage locations in the subsequent two experiments were used as the test set. The results of simultaneously inputting the training and test sets into the SVM are as follows: Figure 4 and Figure 5 As shown.

[0080] Depend on Figure 4 It can be seen that, using the data from the first experiment as the training set, the prediction accuracy for the second experiment is 91.7%; further analysis of the specific details of the prediction results, such as... Figure 4 As shown, one group of samples that were originally in a normal state were incorrectly predicted as having minor damage, while four groups of samples that were originally having minor damage were predicted as having severe damage. Figure 5As shown, using the data from the first experiment as the training set, the prediction accuracy for the third experiment was 85%. Specifically, two groups of samples that were originally in a normal state were predicted as having minor damage, and seven groups of samples that were originally having minor damage were predicted as having severe damage, causing the prediction accuracy to drop to 85%. (Comparison) Figure 4 and Figure 5 The classification results clearly show that the damage to the mechanical turntable exhibits a significant increasing trend over time.

[0081] Compared to traditional methods, the method provided by this invention enables real-time monitoring of damage, thereby allowing for early fault prevention. This invention provides a simpler way to monitor damage to mechanical turntables.

[0082] The present invention has been described by way of example in conjunction with the accompanying drawings. Obviously, the specific implementation of the present invention is not limited to the above-described manner. Any non-substantial improvement made by adopting the inventive concept and technical solution of the present invention, or the direct application of the inventive concept and technical solution of the present invention to other occasions without modification, shall be within the protection scope of the present invention.

Claims

1. A method for monitoring damage to a mechanical turntable based on acoustic emission technology combined with SVM damage classification, characterized in that: Includes the following steps: S1. Install an acoustic emission sensor on the mechanical turntable to be tested; S2. Acquire the acoustic emission waveform collected by the acoustic emission sensor in real time, and extract the AE event based on the original waveform; S3. Calculate the acoustic emission characteristic parameters for each AE event and construct the dataset; S4. Collect AE signals of different damages in different areas of the mechanical turntable, and calculate the acoustic emission characteristic parameters of each AE event to form a training set for different damages. S5. Calculate the acoustic emission characteristic parameters corresponding to the AE signals of different damages detected in the above-mentioned areas, and use them as a test set; S6. Input the above training set and test set into the SVM to obtain the results; S7. Based on the results of SVM, use the accuracy of SVM to monitor the damage to the turntable. In step S7, the number of samples that should have been in normal condition but were predicted as minor damage and the number of samples that should have been minor damage but were predicted as severe damage are counted. Combined with the total number of test set samples, the accuracy is calculated. When the accuracy decreases significantly, it indicates that the mechanical turntable is severely damaged.

2. The mechanical turntable damage monitoring method based on acoustic emission technology combined with SVM damage classification according to claim 1, characterized in that: In step S2, AE events are identified based on preset threshold voltage and duration.

3. The mechanical turntable damage monitoring method based on acoustic emission technology combined with SVM damage classification according to claim 2, characterized in that: In step S2, firstly, data points in the original waveform that exceed the threshold voltage are filtered out, and the positions of data points with an interval greater than the preset duration between two adjacent data points are determined; then, the start and end points of each AE event are recorded, and the duration of the AE event is calculated.

4. The mechanical turntable damage monitoring method based on acoustic emission technology combined with SVM damage classification according to claim 1, characterized in that: In step S3, the acoustic emission characteristic parameters include amplitude, ring count, margin, kurtosis, and centroid frequency.

5. The mechanical turntable damage monitoring method based on acoustic emission technology combined with SVM damage classification according to claim 4, characterized in that: In step S3, the formula for calculating the acoustic emission characteristic parameters is: ； ； ； ； ； ； Among them, X i This represents a time-domain voltage signal, where AM represents amplitude and X represents amplitude. R Mgn represents the root magnitude, and X represents the margin index. max Indicates the maximum value of the signal. Here, denoted by , u represents the signal mean, Kurt represents the kurtosis index, FC represents the centroid frequency, f represents the frequency, and P(f) represents the power spectral density of the signal.

6. The mechanical turntable damage monitoring method based on acoustic emission technology combined with SVM damage classification according to claim 5, characterized in that: In step S3, for the i-th AE event, the dataset is... Where AM represents amplitude, RC represents ring count, Mgn represents margin, Kurt represents kurtosis, and FC represents centroid frequency. This indicates the total number of AE events extracted.

7. A mechanical turntable damage monitoring system based on acoustic emission technology combined with SVM damage classification, characterized in that: The method for implementing the mechanical turntable damage monitoring method based on acoustic emission technology combined with SVM damage classification as described in any one of claims 1-6 includes a damage monitoring system comprising an acoustic emission signal acquisition module, an acoustic emission signal processing module, and a damage monitoring module; wherein: The acoustic emission signal acquisition module includes an acoustic emission sensor, which is used to acquire the AE signal released during the operation of the mechanical turntable under test, and convert the sound signal into an electrical signal; The acoustic emission signal processing module includes a preamplifier and an acoustic emitter. The output of the acoustic emission sensor is connected to the input of the preamplifier via a signal cable. The acoustic emitter is used to filter the signal, perform analog-to-digital conversion, and generate a waveform file for transmission to the host. The damage monitoring module includes a host computer for acquiring the original acoustic emission waveform file and monitoring it according to the steps of the mechanical turntable damage monitoring method based on acoustic emission technology combined with SVM damage classification as described in any one of claims 1-6.

8. The mechanical turntable damage monitoring system based on acoustic emission technology combined with SVM damage classification according to claim 7, characterized in that: The damage monitoring module is also equipped with an indicator light alarm device, which is connected to the host and displays the different degrees of damage to the mechanical turntable under test through different states of the indicator lights.

9. The mechanical turntable damage monitoring system based on acoustic emission technology combined with SVM damage classification according to claim 7 or 8, characterized in that: The acoustic emission sensor is a piezoelectric ceramic sensor with a bandwidth frequency range of 50kHz-400kHz and a resonant frequency of 150kHz; the preamplifier has an amplification factor of 40dB.

Images